Improving Medicare's data on race and ethnicity

Medicare's databases provide a rich source of information about the program's 43 million beneficiaries. These data have played an important role in documenting racial, ethnic, and socioeconomic disparities in health and health care. Because they derive largely from administrative records that have been collected over many years using varying standards, however, they are not fully adequate for monitoring and reducing disparities. The Centers for Medicare & Medicaid Services (CMS) has supported a number of initiatives to improve the quality of its data on race and ethnicity. Yet analyses of 2002 Medicare administrative data show that only 52 percent of Asian beneficiaries and 33 percent of both Hispanic and American Indian/Alaska Native beneficiaries were identified correctly. As CMS moves to reduce disparities, and as researchers strive to explain how and why disparities occur, further improvements in Medicare's data are essential. Health care organizations also need data on the race and ethnicity of the people they serve in order to improve the quality of care for minorities. This brief provides some recommendations for further efforts.